What is a good CCAT score?

What your CCAT score report shows

When you finish the Criteria Cognitive Aptitude Test (CCAT), the hiring test that Criteria Corp sells to employers, the employer gets a score report. You usually don't. Criteria says employers are sent "your raw score (the number of questions you get right) and your percentile"^C1. This is not the Canadian Cognitive Abilities Test that schools use for gifted programs. That test shares the acronym and nothing else.

Criteria's sample score report^C3 shows four things:

1. Raw score. The number of questions you answered correctly, out of 50. Criteria says there is no penalty for wrong answers^C1, so a wrong answer and a blank both add nothing.
2. Percentile. Where your raw score sits against Criteria's reference group. A percentile of 50 means you scored better than half of that group.
3. Three sub-scores. Spatial reasoning, verbal ability, and math and logic, each reported as its own percentile^C4.
4. Suggested score ranges by position. A chart that shows whether your raw score falls inside the range Criteria suggests for each of 42 job families. More on that below, because it is the closest thing to a "passing score" that Criteria publishes.

The raw score is the number everyone talks about. If someone says they got "a 31", they mean 31 correct out of 50.

The average CCAT score is 24

Criteria says it plainly on its candidate page: "the average test taker only gets 24 questions right"^C1. The same page says less than 1% of candidates answer all 50 questions. The test is built that way. Criteria's help documentation says "Most people will not complete the test, and this is by design. The assessment is measuring how quickly someone processes information."^C9

Where does 24 come from? Criteria publishes two reference groups, and they agree on the middle:

The median is the middle score: half the group scored above it, half below. Both groups put it at 24. The standard deviation is how far scores typically sit from the middle. The larger, newer group has a tighter spread, which means more people bunch near 24.

Criteria's sample report gives one exact translation from raw score to percentile: a raw score of 24 "corresponds to a percentile rank of 50"^C3. That is the only point on the curve Criteria publishes. So 25 out of 50, a score people often worry about, is just above the middle of Criteria's reference group. It is not a failing score by any published standard. Whether it is enough depends on the employer.

How your CCAT percentile works

A percentile compares you with Criteria's reference group, not with the other people who applied for your job. Criteria's product page gives the example of a percentile of 45, which means the person "scored better than 45% of the group on which the test was normed"^C2. Your employer may see a different picture in its own applicant pool, and it can sort candidates by raw score, but the percentile on your report is Criteria's.

The same is true of the three sub-score percentiles. A math and logic percentile of 85 means you did better on those questions than 85% of Criteria's reference group did. It does not mean you beat 85% of the people who applied to one company.

You will find full raw-score-to-percentile charts on other prep sites. Criteria doesn't publish one. Those charts are estimates, usually built from a bell curve around 24. They might be close. They might not. We don't print one, because we can't source it, and a wrong percentile is worse than none when you are deciding how hard to prepare.

What you can say with Criteria's own numbers: 24 is the middle. Each point above it puts you ahead of more of the reference group, and because scores bunch near the middle, the first few points above 24 pass more people than the points near the top do.

Is there a passing CCAT score?

No. Criteria does not set a pass mark. Each employer decides what score moves you forward, and some combine it with other tests and interviews. Criteria calls your score "just one of many factors an employer will take into consideration"^C1.

What Criteria does publish is a set of suggested score ranges for job families. Its Information Brief says these ranges "are based on large samples of prior scores from individuals applying for or occupying similar roles and are supported by multiple validation studies"^C4. An employer that links a job to one of these families sees a band on your report showing whether you are in range.

Here are all 42 ranges from Criteria's sample CCAT score report^C3, in the report's own order and wording.

Three things to notice. Every range runs to 50, so the number that matters is the floor. The floors sit between 17 and 24, at or below Criteria's median of 24. Counted by floor: 4 start at 17, 2 start at 18, 8 start at 19, 15 start at 20, 5 start at 21, 6 start at 22, and 2 start at 24. And a range is a suggestion: the Information Brief says employers can create their own score ranges from their own incumbents or applicant pool^C4. An employer hiring for a competitive role can set its own bar well above Criteria's suggested floor, and Criteria doesn't publish those private bars.

So when you read online that "engineers need a 35," ask where the number came from. It may be one employer's policy. It is not in any Criteria document we could find. Criteria's own suggested range for Engineering starts at 22, and for Software Development at 20.

So what is a good CCAT score for you?

A good score is one that clears the bar for the job you applied for. You usually can't see that bar, so work with what you can know:

• Above 24 puts you above the middle of Criteria's reference group.
• At or above the floor for your job family puts you inside Criteria's suggested range, if your employer uses it.
• Your employer may want more. If you can, ask your recruiter whether there is a minimum. Some will tell you. Criteria's support team can't: its FAQ says "Your results are emailed directly to the company for which you applied. The Criteria Support Team does not have access to your results"^C5.

If your test is days away, the useful question is not "what is good" but "where am I losing points." That is something you can find out.

How to read a practice score

A practice score is not a CCAT score. Ours isn't, and no prep site's is. Criteria's percentile compares you with its reference group on its questions. A practice test can only compare you with yourself.

That is still worth a lot, if you read it the right way. When you take a timed 50-question practice test, look at four numbers, not one:

1. How many questions you reached. If you only got to question 30, your score has a ceiling of 30. Pacing is the first thing to fix.
2. Accuracy on the questions you answered. Fifteen right out of 20 attempted is a different problem from 15 right out of 40.
3. Time per question type. Fifty questions in 15 minutes is 18 seconds per question on average. Most people spend far longer on a few types and lose the easy points that come after them.
4. The same pattern twice. One test can be a bad day. Two tests that lose points in the same place are a plan.

On Clever Cat Prep's free practice test, the AI debrief after the test looks at where your points and your time went, by question type. Our full test uses our own mix of question types (18 math, 16 verbal, 10 spatial, 6 logic). That is our simulator's mix, written to match the CCAT's format, not a mix Criteria publishes. Criteria reports math and logic together, so in Criteria's three groups our mix is 24 math and logic, 16 verbal and 10 spatial.

If your practice score is below where you want it, don't read it as a verdict. Read it as a map. Find the two or three question types that cost you the most, practice those, and take another full test. The sample questions by type are a free place to start, and each type has its own page: math, verbal, logic, spatial reasoning and number series.

Can you improve your CCAT score?

Criteria itself acknowledges that practice affects ability-test scores. Its help documentation defines practice effects as when "a person gets higher scores on a test because they've had the opportunity to complete the test multiple times." It limits them two ways: it gives all candidates "the opportunity to familiarize themselves with the test, or with similar tests," and it uses item banks, so a repeat taker "will get a different selection of items"^C10.

Read that carefully. Memorizing questions won't help, because you won't see the same ones. Knowing the question types and how to spend 15 minutes does help, and Criteria wants every candidate to have that familiarity. That is what practice is for.

We can't promise a score, and you should be wary of anyone who does. What practice changes:

• You stop losing time to reading. You recognize a number series or a syllogism in two seconds instead of ten.
• You learn the fast route for the types you know, so you bank more time for the hard ones.
• You learn when to guess. Criteria's own advice: "If you find that you've spent more than 1 minute on a problem, make an educated guess and move on"^C1. There is no penalty for a wrong answer, so never leave a question blank when time runs out.

A full timed CCAT practice test is the closest you can get to test conditions before the day. Take one early, so the second one can measure whether your practice worked.

Three score claims to ignore

"Your CCAT score converts to an IQ." Criteria's report describes the CCAT as measuring "cognitive aptitude, or general intelligence"^C3, but Criteria publishes no IQ conversion. Any table that turns 42 out of 50 into an IQ number was made by someone else.

"Everyone needs a 30 to pass." There is no universal cutoff. Criteria's suggested floors run from 17 to 24, and individual employers set their own.

"The average by industry is X." Criteria publishes suggested ranges by job family, not averages by industry. If a page shows an average for your industry, look for its source.

For the rules around the score, like retakes and whether a second, proctored sitting can happen, see CCAT rules and format. For what the test is and who uses it, see what is the Criteria Cognitive Aptitude Test. If you are also facing the Wonderlic, the scores aren't interchangeable; the CCAT vs Wonderlic page explains why.